<title>Ultrawideband radar evaluation environment for automatic target detection/recognition algorithms</title>

Abstract

Ultra-wideband (UWB) synthetic aperture radar (SAR) is a promising new technology that has the potential for detecting and recognizing targets obscured by foliage or buried beneath the ground. The Army Research Laboratory (ARL) is conducting a significant research effort in Automatic Target Detection/Recognition (ATD/R) algorithms for UWB SAR. The goal of the current ATD/R algorithm research effort is to develop a baseline algorithm approach to accompany newly developed UWB SAR hardware into near-term technology demonstration programs. To accomplish this goal, ARL has assembled a team of researchers from within ARL and from several universities, having expertise in radar, electromagnetics, signal processing and target recognition. Many new and promising ideas currently are being exploited. An evaluation environment was needed to support the integration of those ideas into a coherent algorithm approach. The University of Florida, as part of this project with ARL, has developed a software environment for the development, integration and evaluation of UWB SAR ATD/R algorithms. This paper describes the software environment that serves as the vehicle for integrating the efforts of the various algorithm researchers by providing a set of standards and tools that will enable the various algorithms to be hosted on the same computer system, access a common database, and be evaluated in various combinations against a statistically significant quantity of field-collected data.